Jerusalem in Tønsberg round churches and storytelling Wienberg, …€¦ · JERUSALEM IN TØNSBERG – ROUND CHURCHES AND STORYTELLING Jes Wienberg (Manuscript 4 September 2017)

LUND UNIVERSITY

PO Box 117221 00 Lund+46 46-222 00 00

Jerusalem in Tønsberg

round churches and storytellingWienberg, Jes

Published in:Jerusalem in medieval Scandinavia

2017

Document Version:Early version, also known as pre-print

Link to publication

Citation for published version (APA):Wienberg, J. (2017). Jerusalem in Tønsberg: round churches and storytelling. Manuscript submitted forpublication. In K. A. Aavitsland, & B. Bandlien (Eds.), Jerusalem in medieval Scandinavia: kings, crusaders, andmonastic orders

Total number of authors:1

Creative Commons License:Other

General rightsUnless other specific re-use rights are stated the following general rights apply:Copyright and moral rights for the publications made accessible in the public portal are retained by the authorsand/or other copyright owners and it is a condition of accessing publications that users recognise and abide by thelegal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private studyor research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal

Read more about Creative commons licenses: https://creativecommons.org/licenses/Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will removeaccess to the work immediately and investigate your claim.

1

JERUSALEM IN TØNSBERG – ROUND CHURCHES AND

STORYTELLING

Jes Wienberg (Manuscript 4 September 2017)

1 Discovery Remains of a medieval church were discovered in Tønsberg in Norway in 1877–78, when a new house was built. The identity of the church was easily determined as the Premonstratensian monastery church of Saint Olav. However, the unusual form and size of the church was a puzzling surprise. The remains revealed the largest Romanesque round church in Scandinavia, 27 m in external and 23 m in internal diameter, and the basilical nave had 8 central pillars (fig. 1). Round churches are frequently perceived as enigmatic because of their unusual architecture. The still standing or ruined round churches therefore attract great attention both from professionals and laypeople. Questions are asked and theories are put forward concerning their dating, their function, their social context, type of church chosen and if there was an overall plan. When professionals cannot answer these questions the Knights Templar are free to ride onto the stage and with them the search for secret geometry and hidden treasure. The aim of this lecture/article is to comment critically on the stories we create to explain the round churches, to present my own view on round churches in general and the church in Tønsberg in particular as not being mysterious. 2 Saint Olav’s in Tønsberg Since its discovery the architecture, history and property of Saint Olav’s church and monastery in Tønsberg have been described several times. Most of this is uncontroversial information (e.g. Frölén 1910–11: II, 11-15; Johnsen 1929: 206-32; Lunde 1971; 1993; Wienberg 1991: 38-45, 94-97, 108-11). The controversies take off in a number of competitive interpretations of the function and reconstruction of the round church, on its dating, and on the identification of the initiators and their motives. Saint Olav’s in Tønsberg has been interpreted and reconstructed as a fortified church several times, often inspired by churches on Bornholm: The art historian Anders Bugge presented the church as fortified in an article, where it was drawn with two floors and crenellation by the architect Harald Sund (Bugge 1932; fig. 2). The possibility of a fortified church was also suggested by the medieval archaeologist Øivind Lunde, who perceived its localisation as “strategic” near the entrance to the town from the east (Lunde 1971: 73). The advertising

2

expert Harald Sommerfeldt Boehlke wrote, that the church might have been fortified and drew a building with four floors, with a cross-section reminiscent of the structure of Østerlars on Bornholm (Boehlke 2000: 65-68; 2007: 87 fig. 14). The society “Rundkirkens Venner” (Friends of the Round Church) has released a leaflet with a photomontage made by the illustrator Morten Myklebust, where the round church of Saint Olaf on Bornholm is relocated to present-day Tønsberg (“Rundkirken i Tønsberg” at www.rundkirken.no). This choice again of a model from Bornholm is made despite of this church on Bornholm having a diameter of only about 14 m, half the size of the nave in Tønsberg, and a structure with three floors resting on a central pillar. Most recently, a model in bronze (fig. 3) has been unveiled at the ruin in 2015, where the church is presented as a copy of Nylars on Bornholm, again a church with three floors resting on a central pillar. To relate Saint Olav’s in Tønsberg to crusades is tempting, as a famous chronicle, “Historia de profectione Danorum in Hierosolymam” (The history of the journey by the Danes to Jerusalem) described a rather failed expedition by Norwegians and Danes passing the town; when they finally arrived in Jerusalem, which had fallen in 1187, it was too late for fighting. The chronicle was written by a canon “X” in the 1190s, probably in the Premonstratensian monastery (ScrMin II: 457-92; Krøniker: 117-75; Skovgaard-Petersen 2001; Svenungsen 2016: 105-14). The round church of Tønsberg has been interpreted as a copy or memory of Holy Sepulchre in Jerusalem. The architect Johan Meyer drew the round church as a copy of Holy Sepulchre in Cambridge (Johnsen 1929: 211-12). And the medieval archaeologist Øystein Ekroll has pointed to the inner diameter as a reference to Holy Sepulchre in Jerusalem (Ekroll 2015: 339-40). The round church of Saint Olav must have been built at the latest in 1207, when King Erling Magnusson Steinvegg died in Tønsberg and was buried in the “stonewall” near the altar of the church according to the Baglar Saga (Eirspennill: 457). However, it might have been built earlier as a monastery “Munkelif” (Monk life) was mentioned in 1190 (Sverris Saga: 121), and according to the crusader chronicle the church of Saint Michael at the (later Castle) Mount was supporting Premonstratensian canons in the town in 1191 (ScrMin: II, 473-74; Krøniker: 145-46). It has been suggested by Anders Bugge and Øivind Lunde that the round church was built in the period c. 1160–80 by Earl Erling Ormsson Skakke († 1179) and his son King Magnus Erlingsson (†1184), to legitimize his rule in Viken; the kingdom was taken as a fief of Saint Olav’s at the coronation in 1163 (Bugge 1932: 88-92; Lunde 1993: 15-16; also Ekroll 1997: 169-70). Furthermore it has been proposed that the round church might not have been a monastery church from the beginning, as a stone building has a different alignment from the church (Lunde 1993: 20; also Rasmussen 1968: 552). Since the 19th century it has been presented as a fact that the Knights Templar erected round or polygonal churches (e.g. Frölén 1910–11: 1, 11-15; Krautheimer 1942: 21; Ödman 2005:

3

118-36). In line with this the Norwegian art historian Hans-Emil Lidén wrote in his Norges kunsthistorie (Norwegian Art History) that the church of Tønsberg resembled the churches of the Knights Templar (Lidén 1981: 30-32). Øivind Lunde noticed that both the Temple Church in London and Holy Sepulchre in Cambridge were built by the Knights Templar (Lunde 1993: 12-13). Finally, Harald Sommerstedt Boehlke has in his book Det norske Pentagram (The Norwegian Pentagram), and later also in The Viking Serpent claimed that there was a sacred geometry with lines stretching from Orkney to Gotland, including a line from Trondheim to Tønsberg; the round church of Tønsberg is also reconstructed in a drawing as a copy of the Temple Church in London (Boehlke 2000: 65; 2007: 86 fig. 12). To summarize, we have several interpretations, scientific and/or alternative, but we do not know for sure when precisely the round church was built, if it was a monastery church from the beginning, the identity of the initiator(s) or why it happened. Do these questions and uncertainties turn Saint Olav’s in Tønsberg into an enigma? No! This is a normal situation concerning medieval churches in Scandinavia. Now let us look more closely into the main stories behind the interpretation of Saint Olav’s in Tønsberg and other round churches. 3 Castles of God The idea of fortified or defensive churches in Scandinavia goes back to the 17th and 18th centuries. Churches on Öland in Sweden were described in 1634 as constructed for both worship and defence (Boström 1966: 61, 70-72). And the round churches on Bornholm were described in 1756 as citadels or fortified towers, where the population might take refuge and hide their treasures from pirates and enemies – and from where they themselves might plunder others (Thurah 1756: 52 cf. Wienberg 2004: 36). The concept of defensive churches developed in the 19th century and culminated in the decades around 1900, i.e. in a period when land-based fortifications were a subject of political debate. This was the context for the art historian Hugo F. Frölén, who interpreted all round churches as fortified in his ground-breaking two-volume dissertation Nordens befästa rundkyrkor (The Fortified Round Churches of Scandinavia; Frölén 1910–11). Even when the idea of fortified churches was fiercely attacked back in its heyday as unconvincing when compared to contemporary castles (e.g. Blom 1985; Mowinckel 1928), it remained very popular during the 20th century (Wienberg 2004). It was simply a good story to tell. The idea of what has been called “Castles of God”, is not at all an entirely Scandinavian phenomenon, but an international issue with the best examples being the fortified churches of Transylvania in present-day Romania. Here there is no doubt that they actually were fortified (Harrison 2004; Fabini 2010).

4

The popularity of fortified or defensive churches seems to follow ideological conjunctures over time. In the decades after World War II the question of fortification at churches, as also the investigation of castles, was downplayed. Instead the churches in the Baltic Sea region with upper floors were interpreted as storage for commodities or taxes, as rooms for accommodation or meetings. New concepts were introduced – merchant churches, churches with secular functions, storage churches and multi functional churches (Anglert 1993: 164; Wienberg 2004: 38-41). The downplaying of the defensive church took place despite the tension of the Cold War with an Iron Curtain down through Europe and hot wars in e.g. in Korea and Vietnam. In the present, when Scandinavian countries are participating in “peacekeeping” abroad the concept of defensive churches has been revived, looking for fortified church towers and churchyards (e.g. Bertelsen in Danmarks Kirker IX, 2010: 23–24, 2256-62; Skov 2010). Maybe we have been too naïve in the modern West, ignoring the importance of conflicts in both the present and past (cf. Keeley 1996; Bornfalck Back 2016). The recurrent core examples of the stories about fortification or defence are the four round churches of Bornholm; every year more than 100,000 tourists visit Østerlars (fig. 4), the largest of them. However, the churches on Bornholm are not representative for Scandinavia. They belong to a certain type of multi-functional churches with several floors most common at the Baltic Sea. The four round churches happen to be well preserved because of the relative poverty of the island during the Late Middle Ages, whereas the majority of the eight round churches on Zealand disappeared during Gothic rebuilds (Wienberg 2002: 184-86; 2014: 212-13). 4 Round churches and crusades From written sources it is documented that medieval round churches in Europe were perceived as copies of Holy Sepulchre in Jerusalem (fig. 5) and in a number of cases were erected by returning crusaders as a kind of memorial – e.g. in Senlis in France, in Northampton in England and Eichstätt in Bavaria (e.g. Krautheimer 1942; Kroesen 2000: 12-43; Morris 2005: 223, 230-45; Krüger 2006: 66). The round church as a memorial might be compared to returning Vikings erecting rune stones mentioning distant expeditions or pilgrimages. Thus in Täby in Sweden Estrid erected a rune stone in memory of her husband Östen, who went to Jerusalem but died in “Greece”, i.e. in Byzantium (Sveriges runinskrifter 6: Uppland 136). An incisive example of this view is to be found in Frölén: “By erecting in his homeland a copy of the Holy Sepulchre he might not only receive a confirmation of an already achieved indulgence, but, what was more important, he himself could rest and receive his death masses in a building which – however smaller and simpler

5

– might be said to depict the Holy Sepulchre of Christ himself. However, other motives might have played a part as well. By erecting such a memorial at home an abbot or knight could in the best way strengthen his personal reputation and consolidate the memory of his journey.” (Frölén 1910–11: I, 8). When research into crusades was revived from the 1970s and the definition was broadened to include expeditions in the Baltic Sea and the Iberian Peninsula (cf. Jensen 2000a; Lind et al. 2004) it also renewed the interest in Scandinavian round churches. Thus the round church of Valleberga in Scania (fig. 6) was suggested to have been erected by a crusader not returning from Jerusalem, but from a Northern Crusade (Andrén 1989). Upper storeys in churches were now interpreted as space used as accommodation for travelling knights or pilgrims, for meetings in the Canute Guilds or storage of equipment used in the crusades (cf. Wienberg 2004: 38-40, 43-44). An example of the renewed interest in the Baltic Crusades is by the journalist Jan Eskildsen. He proposes that the round churches on Bornholm and at Kalmar might be dated to the 1120–30s and could have been built as a consequence of the so-called Kalmar expedition, a crusade in 1123 by King Niels of Denmark (he did not show up), King Sigurd Magnusson the Crusader of Norway and Duke Boleslaw III of Poland to Christianize Småland in Sweden. Poland is known to have had many minor round churches. Eskildsen is inspired by the studies of Scandinavian-Polish connections conducted by the art historian Evert Wrangel back in the 1930s, which was interrupted by World War II. He even finds that round churches are built on land in Sweden belonging to the Danish King Valdemar II (Eskildsen 2014: 77-109, 155-57 with fig. p. 84; cf. Wrangel 1933; 1935; Krambs 2014). Recently I have rejected the idea by Frölén and others to narrow down the initiative of round churches to returning crusaders. Many possible church builders actually participated in the crusades. However, an example from Paderborn in Germany demonstrates a need for a more open quest for initiators and their motives. Thus the Bishop of Paderborn, Heinrich II of Werle, was persuaded to erect a round church in Krukenberg in 1126 instead of going on a planned pilgrimage to Jerusalem (Morris 2005: 232-33). I have therefore interpreted the round church as a conspicuous, but at the same time ambiguous, symbol of the crusader ideology (Wienberg 2014: 226-31; cf. Jensen 2000b: 62-65): “Look, I, the church initiator, have already been to Jerusalem or another similar destination, I intend to go soon, I would go if only it was possible – or I support the whole idea of crusades.” The latest contribution by the art historian Kersti Markus is in line with the crusader perspective. Markus perceives the round churches as a “visual culture” in the context of the Danish crusader kingdom during the Valdemarian rule. From written sources, types of churches, choice of building material, iconographic interpretation and the social and historical context she writes a story with a more precise chronology than seen before, relating individual churches to named kings, (arch)bishops and the aristocratic Hvide clan – and also relating the round churches to the political and ecclesiastical development. The Danish round churches are all framed into three main periods: 1) the 1120s with the two-cell round

6

churches of Schleswig (phase 1) and Roskilde influenced by the Polish Piast dynasty. 2) 1171–74 when there was cooperation between King Valdemar I and Archbishop Eskil i.e. after Eskil’s visit to Jerusalem and before he went into exile; Schleswig (phase 2) and the majority of round churches on Zealand e.g. Bjernede (phase 1 in stone). 3) Around 1200, i.e. after the fall of Jerusalem, a number of round churches are built in brick, e.g. Thorsager, Bjernede (phase 2) and also the five-tower cruciform church of Kalundborg by Esbern Snare (†1204). The four round churches of Bornholm are given a relative chronology starting with Østerlars c. 1150 in the context of a local manor, then Saint Olav’s as a fortified round church together with the nearby castle of Hammershus on the initiative of Archbishop Absalon of Lund (†1201), finally Nylars (fig. 7) and Nyker after c. 1200. Absalon and his brother Esbern Snare might have been inspired in their building by a visit to the octagonal chapel in Trondheim in 1188 (Markus 2015). 5 Round churches and the Knights Templar When mentioning crusades it is difficult to avoid the Knights Templar, who have achieved a special position in history, popular and alternative culture as the medieval knights of great power and deep secrets. Nor is it a coincidence that final scenes of Dan Brown’s novel, later turned into a movie, The Da Vinci Code, were located in the round Temple Church in London (Brown 2003). And it is no coincidence that the terrorist Anders Behring Breivik identified himself as a crusader and Knight Templar. We are here dealing with an apparently innocent and entertaining genre of “invented history” filled with conspiracies and esoteric speculations with deep roots in a radical political environment (Andersen 2006: 57-87). Where the Knights Templar appear in popular and alternative culture they are often related to postulated sacred geometries and secrets. Thus the journalist Erling Haagensen has promoted the round churches on Bornholm as built by the Knights Templar in books, websites and movies, some of them together with the author Henry Lincoln. The churches are supposed to have been located according to a complicated sacred geometry and hiding the secrets of the Holy Grail or Templar documents. The churches are interpreted as observatories used by the Templars to measure the circumference of the Earth (e.g. Haagensen & Lincoln 2000). However, except for a few cases, the Knights Templar did not erect round or polygonal churches. It is a factoid wrongly repeated again and again, as pointed out by several people – seemingly in vain (Götz 1968: 289-98; Untermann 1989: 77-81; Naredi-Rainer 1994: 116-37; Morris 2005: 235; Eskildsen 2014: 111-58). For example, Holy Sepulchre in Cambridge was built by three men united in a religious fraternity (Morris 2005: 232-33). And there are no clues whatsoever of any Knights Templar ever been on Bornholm (Wienberg 2002) – and no clues in Tønsberg either. Now let us look more closely at my view of Scandinavian round churches, at Saint Olav’s in Tønsberg after an interlude of 34 years, and at storytelling in general.

7

6 Round churches – rare, but normal The round churches in Scandinavia are exclusively few. According to my present survey covering medieval Scandinavia and including churches with octagonal naves, 34 round churches (fig. 8) are documented. More round churches will be discovered, as happened in Tønsberg in 1877–78 and most recently in Klåstad (fig. 9) in Östergötland in Sweden in 1997 (cf. Hedvall 2007), although the number cannot change dramatically in the future. Thirty-four round churches represent only 0.8 % out of roughly 4400 medieval stone churches in the region (Wienberg 2014: 209-13). Almost the same relationship can be seen in the occurrence of round tower churches, where 16 or 3.8 % are known in a population of about 420 Romanesque church towers in medieval Denmark (Wienberg 2009, cf. 1993: 103-04). The round churches are very few in number, but they are normal in most other respects. When the patron saints are known, they do not deviate from the choice at other medieval churches in Scandinavia – All Saints, Holy Cross, Holy Spirit, James, Lawrence, Martin, Mary, Michael, Nicholas, Olav, Paul and Peter; none of the Scandinavian round churches are known to have been dedicated to the Holy Sepulchre (Wienberg 2014: 210). The size and architecture of the round churches varied from the tiny two-cell church at Orphir in the Orkneys up to the large basilical round church in Tønsberg. The round churches did not deviate from the regional building tradition, whether the churches were normal with one floor, or there was a tradition of two or three floors or even a tradition with fortifications as in the Baltic Sea (Wienberg 2014: 213-15). Their function varied from castle chapels such as Saint Michael’s in Helsingborg to parish churches such as Hagby in Småland and the monastery churches, the Benedictine Saint Michael’s in Schleswig and the Premonstratensian Saint Olav’s in Tønsberg (Wienberg 2014). The round churches in Scandinavia are all Romanesque from the 12th and 13th centuries, but it is only possible to date a handful more precisely, namely Schleswig in Jutland before c. 1140, Bjernede on Zealand 1151–86 and Tønsberg before 1207 (written sources), Nylars on Bornholm after 1202 (coin under main altar) and Voxtorp (fig. 10) in Småland after 1241 (dendrochronology, maybe of a second phase); and all these dates can be and have been disputed (Wienberg 2014: 215-18; also Beseler 1985: 71-72). Finally, from written sources and the local context of runic stones, so-called Eskilstuna monuments, medieval manors or castles, it is probable that round churches were erected by an elite of kings, earls, bishops and noblemen/women, of whom at least some had experiences of crusades in the Baltic or the Mediterranean (Wienberg 2014: 218-21). However, this is no new or surprising observation (e.g. Frölén 1910–11: I, 136-37; Johannsen & Smidt 1981: 53-61; Nilsson 1994: 42-43).

8

To summarize, the Scandinavian round churches were “normal” apart from their elite context and for being round. So why then did the elite erect this rare architecture? 7 Conspicuous round churches The round churches can be described as a “conspicuous architecture” (cf. Wienberg 2014). They were conspicuous in their medieval presence and are conspicuous today. They were meant to attract attention – and they still do. The medieval elite of kings, earls, bishops and noblemen/women chose to erect a conspicuous architecture in their mutual rivalry for status (cf. Johannsen & Smidt 1981: 53) – or to keep the necessary “distinction” from others (cf. Bourdieu 1979). The concept of “conspicuous architecture” is inspired by the sociologist Thorstein Veblen, who introduced the concept of “conspicuous consumption” to describe the lifestyle of the American upper class (Veblen 1899). This concept by Veblen has already been used by archaeologists, for example by Bruce G. Trigger to understand monumental architecture and by Jan Brendalsmo to explain the elitist church building of medieval Trøndelag (Trigger 1990; Brendalsmo 2006: 24-28). Shortly afterwards I also used the concept to understand the Romanesque round church towers of Scandinavia (Wienberg 2009: 110-11). The round churches occurred where the presence of the elite was dense, that is, in landscapes with many monuments and manors. The round churches were also surrounded by richly decorated churches with apses, early towers and galleries. Thus the round church is an appeal for attention in a competitive aristocratic environment. There were several possibilities to receive attention: To build in stone or brick, when these materials were new; to build large, for instance a basilica; to build with a cruciform, round or polygonal church; to add an apse, a central or western tower, a round tower or twin towers; to establish a monastery; and to give rich paintings and inventories. So why choose to erect a monastery with a large basilican round church as in Tønsberg? Why a round building symbolizing Jerusalem? 8 Jerusalem in Tønsberg Every medieval church building represented a new creative situation. Forms and meanings from different buildings were reused in new contexts and had partly new meanings. Inspiration might come from many physical and metaphysical locations. Every building thereby becomes a unique and ambiguous node in an infinite web of influences (Wienberg 2014: 228-29). According to the art historian Richard Krautheimer, prestigious buildings such as Holy Sepulchre in Jerusalem were imitated by selecting minor parts in a “metonymic copying”,

9

where the form was less important than the symbolic content. The explanation for the great variety in architecture was to be found in the way people copied by citing selected elements, in the ability to copy and in the fact that elements were combined in new ways (Krautheimer 1942; cf. Johansen & Smidt 1981: 105-06). When copying Holy Sepulchre you might select the round plan, the ambulatory, the number of pillars or columns, the cupola, the gallery, the absidoles, the aedicule, the dimensions, the liturgy, the dedication or you might only bring back relics from a pilgrimage. The metonymic principle meant that the central fourteen columns and six pillars of Holy Sepulchre might turn up somewhere else copied in the number of twenty, fourteen, eight, six, four or maybe only one. The round church of Saint Olav in Tønsberg was undoubtedly a symbolic copy of Holy Sepulchre in Jerusalem in the basic plan. Melted copper has been found in the centre of the nave; as a hypothesis, this might be remains of an altar representing the “aedicule”, the building covering the tomb of Christ in Holy Sepulchre. When it comes to the reconstruction of the ruin it cannot have looked like the fortified churches with several floors on Bornholm, even though the ground plan of Østerlars has some similarities to a rounded chancel and six pillars in the nave, creating a narrow “oven”; they belong to another type and regional context. Instead one should look for a church with an apse, a rounded chancel and a basilican round nave, preferably with eight pillars supporting a central clerestory. Elements of this can be identified at Holy Sepulchre in Cambridge and Northampton (both eight pillars), as also at the Temple Church in London and Saint Michael’s in Schleswig (both six pillars). Accordingly Morten Myklebust has created a new inspiring photomontage (fig. 11) in 2017 to show how the church of Saint Olav in Tønsberg might look if it is rebuilt in present-day Tønsberg as a copy of Holy Sepulchre in Cambridge. However, the creativity in the process of copying means that there is no exact model out there to be identified resolving all the questions. We do not know the precise dating of the other churches in the town, but they all used the local stone “Tønsbergitt”, including the round church, broadly indicating a common origin in handicraft and time (cf. Brendalsmo & Sørensen 1997). Still, one might say that Saint Olav’s as conspicuous architecture would have to compete with the Saint Michael’s central tower church on the (Castle) Mountain, the Saint Lawrence basilican church with a central tower and twin western towers – and two more ordinary parish churches. If it did not win this competition, it would certainly not be ignored. However, the round church in Tønsberg was not the only representation of Jerusalem in Viken. Every Romanesque apse framing an altar symbolized Holy Sepulchre and its aedicule. If we are looking for location in relation to the expression of power or the past, one might notice that the monastery was established at the foot of the Haugar thing with its two, maybe once three, barrows, where kings were proclaimed – and directly on a burial ground where boat graves have been excavated (Brendalsmo & Molaug 2014: 149).

10

It has been proposed that the church was earlier than the monastery. However, I find it remarkable to construct such a large church, requiring several clerics, if it was not a collegial church or monastery from the very beginning. The stone house belonging to the monastery might simply represent a later phase, as many monasteries were expanded over a long period of time. It might have been built as a so-called “House monastery” (cf. Hill 1992), a burial place and mausoleum for the royal dynasty, although the Bagler Erling Steinvegg was the only king who ended up here (Johnsen 1929: 214-15; Bugge 1932: 95). I have nothing new to add to the discussion of the dating of the round church in Tønsberg. The dating still depends on speculation concerning possible initiators and political motives. However, if we accept another suspect as initiator, the church might belong to the 1190s. 9 Prime suspect and the Bagler faction The prime suspect behind the round church in Tønsberg, Earl Erling Skakke, is known to have been a crusader who reluctantly accepted the Danish King Valdemar I as his royal lord in Viken. Valdemar I visited Tønsberg with a fleet in 1165 and 1168, his son Valdemar II in 1204 (Saxo book 14, chap. 29,18; 38,2; Johnsen 1929: 82-85, 106). The rule in Viken was disputed and uncertain. King Magnus Erlingsson was not the son of a king, but his father Erling was a crusader as also his more famous grandfather King Sigurd the Crusader (Helle 2000; 2003; Svenungsen 2016: 93-95). A large round church referring to the crusader ideology and reminding of crusaders in the dynasty was well chosen as a symbol supporting a faltering rule. Furthermore, it consolidated the Danish kingdom with almost similar basilican round monastery churches at the border to the south in Schleswig and at the border to the north in Tønsberg. However, as the example from Paderborn has demonstrated, the church initiator does not need to have participated in a crusade him- or herself – it might instead be the bishop. In Scania, for comparison, it is known to be Archbishop Eskil of Lund, friend of Bernhard of Clairvaux, who was the initiator of several Premonstratensian monasteries together with the king (Wallin 1961). It is actually mentioned in 1533 by the Oslo Bishop Hans Reff that the monastery of Saint Olav was built and founded by “our ancestors” (DN: X 667; cf. Johnsen 1929: 207; Bugge 1932: 88-90; Lidén 1981: 30-31). Therefore a second suspect might be the Bishop of Oslo, either Helge I (†1190) or maybe more likely his successor Nicholas Arnesson (†1225), the last a half-brother of King Inge Haraldsson the Hunchback (†1161) and a supporter of King Magnus Erlingsson. Bishop Nicholas together with Archbishop Eirik Ivarsson of Nidaros (archbishop 1188–1205, †1213) were the initiators in 1196 in Scania of the “Bagler faction”, named after the bishop’s staff, the “baculus”. The Bagler were striving for power, having Tønsberg at their main base

11

and interacting with the Danish crusader rulers and Absalon, the Archbishop of Lund (Stefánsson 2000; Bagge 2003). After his death Nicholas was succeeded as bishop in Oslo by Abbot Orm from Tønsberg, probably from the Premonstratensian monastery (Johnsen 1929: 131). It is a thrilling thought if Bishop Nicholas might have taken the initiative for the round church and monastery in Tønsberg in the 1190s, because his political ally, the archbishop, was probably the initiator of the erection of an octagonal shrine chapel at Nidaros Cathedral. Because according to the latest investigation by the medieval archaeologist Øystein Ekroll, the octagon must have been erected c. 1200–20 (Ekroll 2015: 347-58). In that case the Bagler faction created two copies of Holy Sepulchre in Jerusalem, both dedicated to Saint Olav, one in the south and one in the middle of the Norway. This happened in a time of crusader mobilization as also seen in the crusader chronicle written either in the Premonstratensian monastery in Tønsberg or in the monastery of Børglum in Denmark (cf. Svenungsen 2016: 107; Skovgaard-Petersen at the conference). The memory of a failed crusade to Jerusalem could be eclipsed by a large material memorial – Jerusalem in Viken. However, the number of suspects might be made greater: Other possible initiators have been mentioned: King Inge the Hunchback (Johnsen 1929: 207), the Danish Archbishop Eskil and King Valdemar I (Svenungsen 2016: 125) and King Sigurd the Crusader (†1130; Jan Brendalsmo in mail 15 Feb. 2017). So an independent dating of the church would be nice, as we are moving here within a wide time frame. 10 Deductive interpretations Examining more closely the different stories about the Scandinavian round churches, we discover that they rest on uncertain ground even when they are told with great conviction leaving no hesitation. The stories are full of biased interpretation, hypothesis and speculation – including my own story, of course. Most of all there is a tendency to “deductive interpretation”, i.e. the overall perspective controls the perception of facts. In the aspiration for stories including all round churches, gaps are filled with hypotheses and contradictions are ignored. The defensive or fortified churches in Scandinavia rest on very little evidence. In fact there is no evidence of defensive churches ever having been attacked or besieged during the Middle Ages in Scandinavia. The few churches known from written sources to have been used for refuge were normal unfortified churches. Similarly, the Scandinavian round churches might be a symbol of Holy Sepulchre in Jerusalem, but there is no evidence relating them directly to the crusades and the principle of “metonymic copying” makes every inquiry into influences arbitrary. The round churches are in several cases dated according to historical interpretations, where independent dating should be preferred: Østerlars should have been erected after 1149,

12

when three-quarters of the island was donated to Archbishop Eskil, and before 1161, when the archbishop went into an exile (Smidt 1935). Valleberga must have been built by Eskil after his return in 1167, but before he retired in 1177 (Svanberg 2002: 347-62) – or in the period 1171–74 (Markus 2015: 18-19). Round churches must have been erected in the 1120s or 30s in Schleswig and Roskilde (Markus 2015: 11-17) as also on Bornholm and at Kalmar (Eskildsen 2014: 77-109). And as we know, the round church of Tønsberg fits well into the framework of political events either 1160–80 or in the 1190s. As in the national history writing of the 19th century, there is a tendency to relate the significant monuments to persons that appear in the written sources: Orphir in the Orkneys and Earl Ragnvald Kolsson, Schleswig and Earl Knud Lavard, Pedersborg and Peder Strangesen, Thorsager and Bishop Peder Vognsen of Århus or his brother and successor Skjalm Vognsen, Roskilde and Magnus Nielsen, Selsø and Saint Olav on Bornholm and Archbishop Absalon, Søborg and Valleberga and Archbishop Eskil, Visby and Bishop Albert of Riga – and finally Tønsberg and Earl Erling Skakke, King Magnus Erlingsson, or Bishop Nicolas Arnesson. In most cases there is no proof, only circumstantial evidence or wishful thinking. The relations might be right, but we actually do not know. The stories are full of peculiarities even when we ignore the Knights Templar, long-distance geometries and hidden treasures: The round churches on Bornholm, as also Voxtorp at Kalmar, which is claimed to be inspired from Poland, all have a three-cell plan, where most Polish churches are two-cell churches. The datings from at least Nylars and Voxtorp are much later than the Kalmar expedition. The assumed two phases at the round churches in Schleswig (cf. Beseler 1985: 71ff; Vellev 1997) and Bjernede are disputed (Frölén 1910–11: II, 16-17; cf. Danmarks Kirker V: 351-62). The Romanesque north portal of Østerlars is interpreted to be either older or later than the round church depending on the expected dating of the church (cf. Smidt 1935; Danmarks Kirker VII: 395-99; Wienberg 1986: 52-53). The church in Kalundborg is normally dated to c. 1200–25, i.e. the context of Peder Strangesen and Ingeborg, the daughter of Esbern Snare (Danmarks Kirker IV: 3096-3100). Hammershus cannot have been built on the initiative of Archbishop Absalon, as it has been dated to the decades around 1300 (cf. Engberg et al. 2015). Absalon and Esbern Snare cannot have been inspired by a visit to Trondheim in 1188, as the octagon was not built yet (cf. Ekroll 2015: 347ff). Finally, the Swedish round churches are located close to rune stones and other social indicators, but the claimed correlation between the property of Valdemar II and clusters of rune stones with special titles or pictures is highly speculative. The deductive interpretations are constructive as working hypothesis waiting for testing by independent methods in the future. Presented as the truth without hesitation, they are doubtful. 11 Round churches and storytelling There is an evolutionary background for the emergence of “Homo narrans” – the

13

storytelling human being. Confronted with the world, we create patterns and tell stories to make our observations and acting meaningful. Thus we cannot see fragments without looking for patterns – patterns that might be our own inventions (Mankell & Vera 2000; Gärdenfors 2006; Boyd 2009). The round churches are embedded in stories focusing on fortification, secular use or symbolic meaning. Stories are told about the need for protection in periods of unrest, of fireproof storage for commodities before the rise of towns and castles, of crusades and pilgrimage to Jerusalem or in the Baltic Sea, stories about kings, bishops and knights – and stories of great planning. Fact and fiction here entangle to create convincing stories, or just good stories. Conscious selections and deselections are made among the few sources. Fragments from the past are connected and the many gaps are filled with qualified guesses. As postmodernists we might believe all stories to be equal, fact or fiction only being a question of perspective. However, if we do not believe in “alternative facts”, we must approach these stories with scepticism and try to distinguish between scientific facts and pure fiction along a graded scale. The stories of the round churches are often very instructive, exciting and entertaining, but in my opinion, this might be an expression of our longing for an enchanted past when living in a modern disenchanted present, not an expression of a past reality. The difficulties in achieving clear-cut answers to questions of dating, social context and motives are common to most medieval churches in Scandinavia and no reason for postulating mysteries. Therefore, there is nothing enigmatic about the round churches. The enigma as a concept might itself be part of the rhetorical way we try to gain attention for our competitive storytelling. Acknowledgements Thanks to Jan Brendalsmo, NIKU in Oslo, and Bodil Petersson (Linnæus University in Kalmar), for constructive comments on the manuscript, to Solveig Borgehammar (Dalby), Martin Hansson (Lund University), and Rikard Hedvall (Arkeologerna, Linköping) for permission to use their photographs, and to Morten Myklebust (Fantasi-Fabrikken, Tønsberg) for permission to publish his new photomontage of Saint Olav’s church. Thanks also to Alan Crozier, Södra Sandby, for language revision. Bibliography Andersen, Dan H. 2006. Skjulte magter? Hemmelige selskaber i 1000 år. Copenhagen: Aschehoug. Andrén, Anders 1989. Bygherre på korståg. Sydsvenska Dagbladet 22 juli 1989. .

14

Anglert, Marit 1993. ’Kyrkorna i Möre’ in Möres kristnande, (ed.) Henrik Williams, Projektet Sveriges kristnande, Publikationer 2, Uppsala: Lunne Böcker, 145-170 Bagge, Sverre 2003. ’Nikolas Arnesson’ in Norsk biografisk leksikon 6, (ed.) Jon Gunnar Arntzene Oslo: Kunnskapsforlaget, 502. Beseler, Hartwig (ed.) 1985. Die Kunstdenkmäler des Landes Schleswig-Holstein, 11, Die Kunstdenkmäler der Stadt Schleswig 3, Kirchen, Klöster und Hospitalen, Berlin: Deutscher Kunstverlag Blom, Otto 1895. ‘Befæstede Kirker i Danmark fra den ældre Middelalder’ in Aarbøger for nordisk Oldkyndighed og Historie 1895, Serie II, volume 10: 1-86 Boehlke, Harald Sommerfeldt 2000. Det norske Pentagram. Kristningen av Norge, 932 til 1152, Tvedestrand: Eutopia Forlag. (English translation: The Viking Serpent. Secrets of the Celtic Church in Norway, their Serpent Worship and Sacred Pentagram Geometry, Bloomington: Trafford Publishing, 2007) Bornfalck Back, Anders 2016. ‘Konflikt i den arkeologiska rekonstruktionen: en pacificerad förhistoria?’ in Fornvännen 111, 2016: 3: 184-191 Boström, Ragnhild 1966. ‘J. H. Rhezelius’ resedagböcker. Några randanmärkningar’ in Konsthistoriska Studier tillägnade Sten Karling, (eds.) Erik Forssman, Brita Linde and Armin Tuulse, Stockholm: Konsthistoriska institutionen, Kungliga universitetet, 59-78 Bourdieu, Pierre 1979. La distinction. Critique sociale du judgement, Le sens commun, Paris: Les Èditions de Minuit Boyd, Brian 2009. On the Origin of Stories. Evolution, Cognition, and Fiction, Cambridge (Mass.): Harvard University Press Brendalsmo, Jan 2006. Kirkebygg og kirkebyggere. Bygherrer i Trøndelag ca. 1000-1600, Perspektiv 17 x 24 no 2006: 30, Oslo: Unipub forlag (Dissertation 2001) Brendalsmo, Jan and Molaug, Petter B. 2014. ’To norske byer i middelalderen – Oslo og Tønsberg’ in Collegium Medievale 2014: 134-199 Brendalsmo, A. Jan and Sørensen, Rolf 1997. ’Kvader i sentrum. Håndverksmessige og geologiske forutsetninger for bygging av steinkirker i Vestfold i tidlig middelalder’ in Hikuin 22: 77-94 (Officially, but incorrect year of print as 1995) Brown, Dan 2003: The Da Vinci Code, New York: Doubleday

15

Bugge, Anders 1932. ‘Nordens største rundkirke. Olavsklostret i Tønsberg’ in Foreningen til norske fortidsminnesmerkers bevaring, Årsberetning 1931: 87-97 Danmarks Kirker Iff, Copenhagen: Nationalmuseet, 1933ff Diplomatarium Norvegicum, Oldbreve til Kundskab om Norges indre og ydre Forhold, Sprog, Slægter, Sæder, Lovgivning og Rettergang i Middelalderen, I-XXIII, Norsk historisk Kjeldeskrift-institutt, Christiania/ Oslo: P. T. Mallings forlagsboghandel/ Riksarkivet, 1849-2011 Eirspennill. Am 47 fol. Nóregs konunga sogur. Magnus gódi – Hákon gamli, (ed.) Finnur Jonsson, Kristiania: Den norske historiske kildeskriftskommission, 1916 Ekroll, Øystein 1997. Med kleber og kalk. Norsk steinbygging i mellomalderen 1050-1550, Oslo: Det Norske Samlaget Ekroll, Øystein 2015. The Octagonal Shrine Chapel of St Olav at Nidaros Cathedral. An Investigation of its Fabric, Architecture and International Context, Doctoral theses at NTNU 2015: 294, Trondheim: NTNU, Norwegian University of Science and Technology Engberg, Nils, Etting, Vivian, Frandsen, Jørgen, Sonne Nielsen, Finn Ole and Wille-Jørgensen, Dorthe 2015. ‘Hammershus – korsfarer- eller ærkebispeborg?’ in Nationalmuseets Arbejdsmark 2015: 134-167 Eskildsen, Jan 2014. Bornholm og Østersøen 1060-1140. Nedslag i en historisk periode fra før den danske kirke organiserede sig, til der blev planlagt kirkebyggeri, Rønne: Bornholms Tidende Fabini, Hermann 2010. The Church-Fortresses of the Transylvanian Saxons, Sibiu: Monumenta Frölén, Hugo F. 1910-11. Nordens befästa rundkyrkor. En konst- och kulturhistorisk undersökning med 270 bilder, I-II, Stockholm: Lars Frölén Gärdenfors, Peter 2006. Den meningssökande människan, Stockholm: Natur och Kultur Götz, Wolfgang 1968. Zentralbau und Zentralbautendenz in der gotischen Architektur, Berlin: Gebr. Mann Verlag Haagensen, Erling and Lincoln, Henry 2000. The Templars’ Secret Island. The Knights, the Priest and the Treasure, Moreton-in-Marsh: The Windrush Press Harrison, Peter 2004. Castles of God. Fortified Religious Building of the World, Woodbridge/ Rochester: The Boydell Press

16

Hedvall, Rikard 2007. ’Kyrkorna i Klåstad’ in Människors rum och människors möten. Kulturhistoriska skisser, (ed.) Anders Perlinge, Stockholm: Berit Wallenbergs Stiftelse, 143-164 Helle, Knut 2000. ’Erling Ormsson Skakke’ in Norsk biografisk leksikon 2, (ed.) Jon Gunnar Arntzen, Oslo: Kunnskapsforlaget, 501-502 Helle, Knut 2003. ’Magnus 5. Erlingsson’ Norsk biografisk leksikon 6, (ed.) Jon Gunnar Arntzen, Oslo: Kunnskapsforlaget, 212. Hill, Thomas 1992. Könige, Fürsten und Klöster. Studien zu den dänischen Klostergründungen des 12. Jahrhunderts, Kieler Werkstücke, Reihe A, Beiträge zur Schleswig-holsteinischen und skandinavischen Geschichte 4, Frankfurt am Main/ Bern/ New York/ Paris: Peter Lang Jensen, Kurt Villads 2000a. ‘Temaer i korstogshistorien – et historiografisk rids’ in Den jyske historiker 89, Krig, korstog og kolonisering, Århus, 8-29 Jensen, Kurt Villads 2000b. ‘Danmarks som en korsfarerstat’ in Den jyske historiker 89, Krig, korstog og kolonisering, Århus, 48-67 Johannsen, Hugo and Smidt, Claus M. 1981. ’Kirkens huse’ in Danmarks arkitektur, (ed.) Hakon Lund, Copenhagen: Gyldendal, 9-197 Johnsen, Oscar Albert 1929. Tønsbergs historie I, Middelalderen, Oslo: Gyldendal Norsk Forlag Keeley, Lawrence H. 1996. War before civilization. The myth of the peaceful savage, New York/ Oxford: Oxford University Press Krambs, Karsten 2014. Valdemar IIs godser i Swecia. (Article on academia.edu) Krautheimer, Richard 1942. ’Introduction to an ”Iconography of Medieval Architecture”’ in Journal of the Warburg and Courtauld Institutes V: 1-33. (Reprint in: Studies in Early Christian, Medieval and Renaissance Art, London/ New York: University of London Press and New York University Press, 1969, 115-150) Kroesen, Justin E. A. 2000. The Sepulchrum Domini Through the Ages. Its Form and Function, Liturgia condenda 10, Leuven/ Paris/ Sterling (Virginia): Peeters Krüger, Jürgen 2006. ’Die Grabeskirche in Jerusalem und ihre Nachbauten im 11. und 12. Jahrhundert’ in Canossa 1077. Erschütterung der Welt. Geschichte, Kunst und Kultur am Aufgang der Romanik, I, (eds.) Christoph Stiegemann and Matthias Wemhoff, München: Hirmer, 498-511

17

Krøniker fra Valdemarstiden. (Ed.) Jørgen Olrik, Copenhagen: Selskabet for historiske Kildeskrifters Oversættelse, 1900-01 Lidén, Hans-Emil 1981. ‘Middelalderens steinarkitektur i Norge’ in Norges kunsthistorie 2, Høymiddelalder og hansa-tid, (ed.) Knut Berg, Oslo: Gyldendals Norsk Forlag, 7-125. Lind, John H., Jensen, Carsten Selch, Jensen, Kurt Villads and Bysted, Ane L. 2004. Danske korstog – krig og mission i Østersøen, Copenhagen: Høst & Søn Lunde, Øivind 1971. ’Premonstratenserklosteret i Tønsberg. St. Olavs kirke – Nordens største rundkirke’ in Vestfold-Minne 1971: 68-85 Lunde, Øivind 1993. ‘Premonstratensernes kloster i Tunsberg – kirken og klosteranlegget’ in Seminaret “Kloster og by” 11.-13. november 1992 – omkring Olavsklosteret, premonstratenserordenen og klostervesenet i middelalderen, (eds.) Jan E. G. Eriksson and Kari Schei, Tverfaglige seminarer i Tønsberg 1, Tønsberg: Tønsberg Bibliotek & Riksantikvaren, Utgravningskontoret for Tønsberg, 9-22 Mankell, Henning & Vera, Yvonne 2000. Homo narrans – the story telling man: a conversation, Stockholm: Millenniekommitén Markus, Kersti 2015. ’Kuidas propageerida ristisõda visuaalselt? Taanlaste ristisöjast Eestisse 12. sajandi lõpus ja. 13. sajandi algul sakraalse ehituskunsti vahendusel’ in Kunstiteaduslikke Uurimusi 2015, 24: 3-4: 7-48. (Summary: How to Propagate a Crusade Visually? The Danish Crusade in Estonia in the Late 12th Century and Early 13th Century in the Light of Sacral Architecture) Morris, Colin 2005. The Sepulchre of Christ and the Medieval West. From the Beginning to 1600, Oxford: Oxford University Press Mowinckel, Rolf 1928. ‘Har vi hat fæstningskirker her i landet i middelalderen?’ in Norsk teologisk tidsskrift 29, serie 3, vol 9, no. 1-2: 80-99 Naredi-Rainer, Paul von 1994. Salomos Tempel und das Abendland. Monumentale Folgen historischer Irrtürmer, Köln: Dumont Buchverlag Nilsson, Ing-Marie 1994. Rundkyrkor i Norden, Master thesis in medieval archaeology, spring 1994, Lund: Department of Archaeology, Lund University (Stencil) Rasmussen, Jørgen Nybo 1968. ‘Præmonstraneserordenen’ in Kulturhistorisk leksikon for nordisk middelalder fra vikingetid til reformasjonstid 13, 550-554 Saxo Grammaticus, Gesta Danorum – Danmarkshistorien, 1-2. (Eds.) Karsten Friis-

18

Jensen and Peter Zeeberg, Copenhagen 2005: Det danske Sprog- og Litteraturselskab and Gads Forlag ScrMin: Scriptores Minores Historiæ Danicæ medii ævi II. (Ed.) M. Cl. Gertz, Copenhagen: Selskabet for udgivelse af Kilder til dansk Historie, 1922 Skov, Hans 2010. ’Kirkeborge’ in Skalk 2010: 2: 25-30 Skovgaard-Petersen, Karen 2001. A Journey to the Promised Land. Crusading Theology in the Historia de profectione Danorum in Hierosolymam (c. 1200), Copenhagen: Museum Tusculanum Press Smidt, C. M. 1935. ’Østerlarsker paa Bornholm. Nye Undersøgelser’ in Fra Nationalmuseets Arbejdsmark 1935: 59-69 Stefánsson, Magnús 2000. ’Eirik Ivarsson’ in Norsk biografisk leksikon 2, (ed.) (Ed.) Jon Gunnar Arntzen, Oslo: Kunnskapsforlaget, 438-439 Svanberg, Jan 2002. ’En dopfunt och ett brev från 1100-talet. Investiturstriden speglad i Vallebergafuntens bildprogram, in Ny väg till medeltidsbreven. Från ett medeltidssymposium i Svenska Riksarkivet 26-28 november 1999, (eds.) Claes Gejrot, Riger Andersson and Kerstin Abukhanfusa, Skrifter utgivna av Riksarkivet 18, Stockholm: Riksarkivet, 339-366 Svenungsen, Pål Berg 2016. Norge og korstogene. En studie av forbindelsene mellom det norske riket og den europeiske korstogsbevegelse, ca. 1050-1380. (Dissertation, University of Bergen, pdf) Sveriges runinskrifter 1ff. Stockholm: Kung. Vitterhets Historie och Antikvitets Akademien, 1900ff Sverris Saga etter Cod. AM 327 4°. (Ed.) Gustav Indrebø, Norske historiske kildeskriftsfonds skrifter 46, Kristiania: Dybwad, 1920 Thurah, Lauritz de 1756. Omstændelig og tilforladelig Beskrivelse, Over den i Øster-Søen liggende, Under Det kongelige Danske herredømme Blomstrende navnkundige Øe Bornholm, Og Den ei langt derfra anlagde fortreffelige Fæstning Christiansøe…. Copenhagen: Nicolaus Møller Trigger, Bruce G. 1990. ‘Monumental architecture: a thermodynamic explanation of symbolic behaviour’ in World Archaeology 22: 2: 119-132 Untermann, Matthias 1989. Der Zentralbau im Mittelalter. Form, Funktion, Verbreitung, Darmstadt: Wissenschaftliche Buchgesellschaft

19

Veblen, Thorstein 1899. The Theory of the Leisure Class. An Economic Study in the Evolution of Institutions, New York: The Macmillian Company Vellev, Jens 1997. ’Rundkirken Skt. Mikaelis i Slesvig. Arkæologisk undersøgelse og rekonstruktion af en gådefuld bygning’ in Hikuin 22: 47-58 (Officially, but incorrect year of print as 1995) Wallin, Curt 1961. ‘Ärkebiskop Eskil som klosterstiftare’ in Scandia. Tidskrift för historisk forskning 27: 1: 217-234 Wienberg, Jes 1986. ’Bornholms kirker i den ældre middelalder’ in Hikuin 12: 45-66. (Also printed in Bornholmske Samlinger 1989, Serie III, vol. 3: 9-28) Wienberg, Jes 1991. De kirkelige institutioner i middelalderens Tønsberg, Arkeologiske rapporter fra Tønsberg 6, Tønsberg: Riksantikvaren, Utgravningskontoret for Tønsberg (Master thesis of medieval archaeology, Aarhus University, 1983) Wienberg, Jes 1993. Den gotiske labyrint. Middelalderen og kirkerne i Danmark, Lund Studies in Medieval Archaeology 11, Stockholm: Almqvist & Wiksell International Wienberg, Jes 2002. ’Mellem viden og vrøvl – Bornholms mystiske kirker’ (with an appendix by Jørn Schmidt) in Bornholmske Samlinger 2002, Serie III, vol. 16: 175-190 Wienberg, Jes 2004. ’Fortresses, Storehouses and Symbols – ambiguous churches of the Baltic Sea’ in Der Ostseeraum und Kontinentaleuropa 1100-1600. Einflußsnahme - Rezeption – Wandel, (eds.) Detlef Kattinger, Jens E Olesen and Horst Wernicke, Culture Clash or Compromise VIII, Schwerin: Thomas Helms Verlag, 35-50. (Also printed in META. Medeltidsarkeologisk tidskrift 2000: 4: 26-58 and in Bornholmske Samlinger 2003, Serie III, vol. 17: 9-35) Wienberg, Jes 2009. ’Romanske runde kirketårne – et skandinavisk perspektiv’ in Hikuin 36: 101-120 (Also printed in Zwischen Fjorden und Steppe. Festschrift für Johan Callmer zum 65. Geburtstag, (eds.) Claudia Theune, Felix Biermann, Ruth Struwe and Gerson H. Jeute, Internationale Archäologie, Studia honoraria 31, Rahden, Westf.: Verlag Marie Leidorf, 2010, 515-521) Wienberg, Jes 2014. ‘Iøjnefaldende arkitektur. Nordens middelalderlige rundkirker’ in Kuml, Årbog for Jysk arkæologisk Selskab, 2014: 205-244. (English version in press: ‘Conspicuous Architecture – Medieval Round Churches in Scandinavia’ in New Visits to Old Churches – Sacred Monuments and Practices in the Baltic Sea Region, (eds.) Janne Harjula et al., Cambridge: Cambridge Scholars Publishing) Wrangel, Ewert 1933. ’Sverige och Polen. En förberedande undersökning om en ny linje i den äldre medeltidens konsthistoria’ in Tidskrift för konstvetenskap 17: 3: 49-76

20

Wrangel, Ewert 1935. ’Skandinaviska förbindelser med de västslaviska folken under den äldre medeltiden – rundkyrkorna’ in Tidskrift för konstvetenskap 18: 3-4: 103-121 Ödman, Anders 2005. Skeingeborg – borgen som Saxo glömde, Norra Skånes Medeltid 4, Report Series 90, Lund: Department of Archaeology and Ancient History Homepages: Rundkirkens venner: www.rundkirken.no Jes Wienberg ([email protected]) is professor at the Department of Archaeology and Ancient History, Lund University, Sweden. Abstract: Remains of a basilican round church were discovered in 1877–78 in Tønsberg, Norway, and identified as the Premonstratensian monastery church of Saint Olav. The round church has been interpreted as fortified with several floors using Nylars on Bornholm in Denmark as a model – or as symbolic copy of Holy Sepulchre in Jerusalem, although Templar churches have also been mentioned as models. The church of Saint Olav is normally dated to the period c. 1160–80 and supposed to have been built on the initiative of Earl Erling Skakke and his son King Magnus Erlingsson. The round church of Tønsberg is discussed in the article relation to fortification, crusades, the Knights Templar and an overview of the 34 known Scandinavian round churches. The article argues that the Scandinavian round churches were normal in most respects, except for their deviating architecture. The round churches are interpreted as a “conspicuous architecture”, which were meant to attract attention for the crusader ideology and lend prestige to the initiator(s). Saint Olav in Tønsberg may have looked like Holy Sepulchre in Cambridge, although every building process combined different models, thereby created something new. A photomontage by Morten Myklebust relocates Holy Sepulchre in Cambridge to present-day Tønsberg. Furthermore the article proposes that Bishop Nicholas might have taken the initiative for the round church and monastery in Tønsberg in the 1190s. The bishop was an ally of the archbishop of Nidaros/ Trondheim, who was the initiator of a famous octagonal shrine chapel at his cathedral. Keywords: Round church, Jerusalem, Crusade, Knights Templar, conspicuous architecture, storytelling Illustrations Fig. 1. Round church of Saint Olav in Tønsberg, Norway. Photo: Jes Wienberg, August 2016.

21

Fig. 2. Reconstruction of Saint Olav’s in Tønsberg as a fortified church by Harald Sund (Bugge 1932: 87). Fig. 3. Model of Saint Olav’s in Tønsberg as a copy of Nylars on Bornholm. Photo: Jes Wienberg, August 2016. Fig. 4. The church of Østerlars on Bornholm, Denmark. Photo: Jes Wienberg, July 2009. Fig. 5. Holy Sepulchre in Jerusalem. Photo: Solveig Borgehammar, July 2011. Fig. 6. The church of Valleberga in Scania, Sweden. Photo: Martin Hansson, August 2013. Fig. 7. The church of Nylars on Bornholm, Denmark. Photo: Jes Wienberg, August 2011. Fig. 8. Overview of the medieval round churches of Scandinavia (Wienberg 2014: 210). Fig. 9. The excavation of the round church of Klåstad in Östergötland, Sweden. Photo: Rikard Hedvall, October 2000. Fig. 10. The church of Voxtorp in Småland, Sweden. Photo: Jes Wienberg, July 2009. Fig. 11. Photomontage with Holy Sepulchre in Cambridge relocated to present-day Tønsberg by Morten Myklebust, 2017.